Tim Ellison wrote: > Salikh Zakirov wrote: >> The attached patch series explore the idea of making portlib mostly >> independent >> of threading library. It changes portlib to use OS services (via thrdsup.h >> macro wrappers). I have only tried it on Windows, and would like to have some >> feedback before expending more effort on Linux version. > > [I'm allergic to mailing list attachments ... would much prefer a link > to a combined patch in JIRA than have you fill-up hundred's of people's > inboxes]
I would file a JIRA if I had something that is likely to be committed. The attached patch was just an extended version of proposal to discuss. Somehow, JIRA with "discussion patches" leave some uneasy feeling about it. > We already said that going straight to the OS wasn't a great idea > though. If the VM is going to provide monitoring and management of > thread and locks then it needs to see the function calls go by. >> With these patches applied, the only threading functions called from portlib >> are hythread_attach() and hythread_detach(). > > While I agree that those are the only ones that the VM is forced to care > about, we need a solution that allows the VM to care about them all. > > <more stuff snipped, since it depends on agreeing this fundamental approach> I see your point, and I am not opposing it. I would like to see the patch that satisfies this requirement, while making thread library truly VM-specific. -- Salikh Zakirov
